Journal of Voice

Vol. 4, No. 3, pp. 238-249

1990 Raven Press, Ltd., New York
Clinical Applications of Electroglottography
Peter Kitzing
Department of Phoniatrics, University Ear-Nose-Throat Clinic, Mal m6 General Hospital, Lund University, Sweden
Summary: Electroglottography (EGG) is a method to monitor the vibrations of
the vocal folds by measuring the varying impedance to a weak alternating
current through the tissues of the neck. The paper is an attempt to give a
state-of-the-art report of how electroglottography is used in the clinic. It is
based on a search of the pertinent literature was well as on an inquiry to 17 well
known specialists in the field. The EGG techniques are described and limita-
tions to the method are pointed out. Attempts to document voice quality by
EGG are recognized and computerized methods to obtain information about
vibratory perturbations and/or the vibratory frequency of the vocal folds are
described. The author' s personal conclusion is that the EGG signal is espe-
cially well suited for measurements of the glottal vibratory period. In the clinic
such measurements are useful for periodicity analysis, as a basis for recording
intonation contours, and to establish the characteristics of the voice funda-
mental frequency. Key Words: Electroglottography--Electrolaryngography--
Phonat i on--Voi ce quality--Voice pi t ch--Range of speaking voi ce--Voi ce
therapy--Vocal biofeedback.
El ect r ogl ot t ogr aphy ( EGG) (1) or el ect r ol ar yn-
gogr a phy (2) is an e nt i r e l y noni nva s i ve , eas y-
t o- handl e me t hod of moni t or i ng l ar yngeal vi bra-
t i ons, i mper vi ous t o ambi ent noi se, and yi el di ng
seemi ngl y st r ai ght - f or war d i nf or mat i on about t he
vi br at or y behavi or of t he vocal fol ds duri ng phona-
t i on. The i nst r ument at i on is easi l y avai l abl e, ei t her
cust om- bui l t or as one of t he at l east t hr ee commer -
cially avai l abl e devi ces. No wonder , t hen, t hat t he
me t hod has a t t r a c t e d gr eat i nt er es t f r om ma ny
voi ce cl i ni ci ans and t hat r epor t s on clinical applica-
t i ons of EGG are abundant . On t he ot her hand,
EGG is f ar f r om bei ng gener al l y accept ed in clinical
wor k on voi ce di sor der s, and t her e is still i nt ense
wor k goi ng on, ai med at t he devel opment and re-
Addr ess cor r espondence and r epr i nt request s t o Dr. P. Kitzing
at De pa r t me nt of Phoni at r i cs , Uni ver s i t y Ear - Nos e- Thr oat
Clinic, Mal m6 Gener al Hospi t al , Lund Uni ver si t y, S 21401
Mal mr , Sweden.
Paper pr esent ed at t he 18th Annual Symposi um: Care of t he
Professi onal Voi ce, Philadelphia, Pennsyl vani a, June 4-9, 1989.
f i nement of t he EGG me t hod t o make it mor e suit-
abl e f or clinical and r es ear ch use.
Pr i nci pal l y, EGG is a me t hod t o meas ur e t he re-
si st ance or , mor e pr eci s el y, t he i mpe da nc e t o a
we a k al t er nat i ng cur r ent t hr ough t he t i ssues of the
ne c k. Th e s e ha ve mo d e r a t e l y good conduct i ve
qual i t i es, wher eas ai r is an e xt r e me l y poor conduc-
t or . Ther ef or e, duri ng t he vi br at or y cycl e of t he vo-
cal fol ds t he i mpedance is i ncr eas ed whe n t he glot-
tis is opened. Duri ng t he cl osur e of t he glottis, the
el ect r i cal i mpedance dr ops as t he r e is i ncreased
cont act be t we e n t he vocal fol ds. I t is a common
not i on t hat t he dut y cycl e of t he el ect r ogl ot t ogr am
r epr es ent s t he size of t he ar ea of cont act bet ween
t he vocal fol ds, a hypot hes i s t hat has be e n experi-
ment al l y cor r obor at ed by Sc he r e r et al. (3). For a
det ai l ed descr i pt i on of t he EGG me t hod and a thor-
ough i nt e r pr e t a t i on of t he EGG wa v e f o r m, the
r e a d e r is r e f e r r e d t o t he ar t i cl es by Ba k e n (4)
and Ti t ze (5), r es pect i vel y, i mmedi at el y preceding
t he pr es ent paper at t he 18th Sympos i um: Care of
238
ELECTROGLOTTOGRAPHY: CLINICAL APPLICATIONS 239
the professional voice, Philadelphia 1989. On t he
same occasi on, a new EGG was pr esent ed by
Rothenberg (6).
This state-of-the-art report is based on a com-
puter search of the pert i nent literature as well as on
inquiry responses from 17 well-known specialists in
the field (Baer, Chevrie-Muller, Childers, Dejonck-
ere, Fer r er o, Four ci n, FrCkj~er-Jensen, Hacki ,
Hanson, Holmberg, Jent zsch, Lecluse, Pedersen,
Reinsch, Schutte, Sopko, and Unger), whose valu-
able collaboration is her eby gratefully acknowl-
edged. The present aut hor' s own experience of glot-
tography dates back to 196l, starting with photo-
glottography under t he guidance of Sonesson (7).
The report is not organized according to the dif-
ferent available EGG met hods, as is rather com-
mon. Instead it will start from clinically used crite-
ria of vocal function, and it will try to show how
they can be document ed by EGG. The criteria to be
studied are register, quality, intonation, roughness,
and pitch. The last t wo sections of the report will
deal with the use of EGG in certain diagnoses and in
therapy. But, first, t here will be a short survey of
the reported techniques for using the EGG signal,
and some pitfalls and shortcomings of the met hod
will be pointed out.
To fit into the given frame of time and space, t he
material has been condensed extremely, and, of ne-
cessity, the report will remain superficial on many
issues. Therefore, the reader is strongly urged to
consult the ci t ed references for adequate informa-
tion on points of special interest.
EGG TECHNIQUES
The information from EGG can be processed in
many different ways. The generally used and most
obvious met hod seems to be subjective visual eval-
uation of the wave shape (Lx) (8). To get objective
data, durational and amplitude measurement s can
be carried out on different parts of the glottographic
curve by aid of a ruler, with one commonl y used
measure being the time of glottal closure (or closed
phase). As will be discussed under the heading of
voice quality, t hese dat a oft en are parameterized by
COmp0ting so- cal l ed quot i ent s, like t he open-
quotient, i.e., the time of open glottis divided by t he
entire period time. The measuring and computing
process may be comput eri zed (9). Computers may
also be used to differentiate the electroglottogram,
with the resulting curves showing the velocity of
change of the vocal fold cont act area instead of only
the amount (10-13). By plotting (the inverted) elec-
troglottographic period time along a time axis, into-
nation curves (Fx) (8) can be obtained. Large cycle-
by-cycl e variations can al ready be seen in such
curves, but t hey are mor e clearly demonst rat ed in
scatterplots [Cx or digram distributions (8,14) or bi-
histograms (15,16)], wher e the period of one vibra-
t ory cycl e is plotted against t he next. The use of
comput eri zed statistics of period measurement s is
anot her way to handle electroglottographical data.
In this way, measures of central t endency, like t he
mean, the median, or t he mode, as well as measures
of the range can be obtained. The result can be
shown graphically, as t he distribution of frequen-
cies [or F x histogram, according to Fourci n (8)].
Provided that aperiodic or ext remel y low frequen-
cies are excluded from t he computations, t he men-
tioned statistics are reasonabl y represent at i ve of
the mean pitch and range of the voice (17) [(Fx his-
togram of second and third order (8); t runcat ed glot-
tal frequency analysis (18)].
With growing i nsecuri t y as to t he reliability of t he
EGG signal, a number of authors r ecommend that it
be combined with ot her met hods like videostrobos-
copy (19-23), phot ogl ot t ography (24-28), i nverse
fi l t ered flow gl ot t ography (29), xer or adi ogr aphy
(30), and phonet ograms (31).
Recent l y, Titze et al. (32-34) have descri bed an
interesting met hod of treating EGG data alone or in
combi nat i on with phot ogl ot t ograms by mat chi ng
t hem with simulated waveforms of a comput er vi-
brat ory mode of the glottis. Thei r program is called
gl ot t al i magi ng by pr ocessi ng ext er nal signals
(GLIMPES). It can generat e a certain number of
phonat ory paramet ers, of whi ch the so-called ab-
duction quotient has been shown useful in describ-
ing voice quality along t he sequence "br eat hy, nor-
mal, pr essed. "
LIMITATIONS OF THE EGG METHOD
The fact that it is usually very easy to obtain EGG
curves may weaken awareness of the limitations of
t he met hod and of possible pitfalls when interpret-
ing t he results. The pl acement of the electrode~ in
front of the thyroid cartilage and the di st ance be-
t ween t hem is rat her critical (35), as is also t he low-
ering of the skin-electrode electrical resi st ance by
keeping the electrodes clean, lubricated with con-
ductive paste, and firmly at t ached to t he skin. Es-
Journal of Voice, Vol. 4, No. 3, 1990
240 P. KI T Z I NG
pecially in female subjects, it may be impossible to
obtain an EGG registration, most often because of
abundant subcut aneous soft t i ssue (36--38). The
generally smaller anatomical dimensions of the fe-
male larynx than the male may be another cause of
insufficient EGG amplitudes. In this connection,
one may recall that only ~1% of the total imped-
ance through the neck is modul at ed by the glottal
vibrations. Artifacts may be introduced into the sig-
nal by head movement s or vertical shifts of the lar-
ynx (39,40). Compensating electric filtering and au-
tomatic gain control may also introduce distortion
in the signal.
One cause of artifact, put forward for discussion
by Smith (41), is that the glottographic signal could
be acoustically dependent on the tissues acting as a
kind of mi crophone. This possi bi l i t y, however ,
seems to have been ruled out by a number of dif-
ferent experimental results, such as excitation of
the vocal tract by an el ect rol arynx causing no EGG
(1), phonat i on in helium causing no EGG changes in
spite of drastical changes of the sound pressure
waves (42), and the withdrawal of an electrically
isolating pol ymer strip bet ween the vocal folds,
causing an increase of the amplitude (43). Besides,
changes of mechanical cont act bet ween the vocal
folds cause changes of glottal impedance even in the
absence of phonation, as can be easily seen, e.g., in
glottograms published by the present author (27).
To concl ude this section, some caution against a
t oo uncritical use of the EGG seems appropriate.
An all-too-naive interpretation of EGG data may
pr ove detrimental, especi al l y in clinical applica-
tions. The user should al ways be aware that the
electroglottogram does not represent vibrations of
the vocal folds as such, but variations in the area of
cont act bet ween them. And even this is only an
approximation, as the cause of glottographically de-
t ect ed i mpedance changes still are not known in
every detail, nor is the rel evance of vocal fold con-
t act area variations to the acoust i c voi ce signal.
There may be evi dence for a reasonabl y good cor-
r el at i on b e t we e n t he vocal f ol d c ont a c t ar ea
changes and the electroglottograms in a number of
publications, one of the most impressive being the
detailed experiments with anesthetized dogs by the
Uni versi t y of California at Los Angeles (UCLA)
group (44). On the ot her hand, in this connect i on it
seems impossible not to mention the latest evalua-
tion by Childers (45, p. 21), an eminent authority on
the physiological analysis of the el ect rogl ot t ogram,
that it "remai ns a poorl y under st ood tracking de.
vice, in its present f o r m. . , not capabl e of contrib.
uting much to clinical diagnosis and t reat ment of
voi ce di sor der s" (cf. also refs. 46--49).
EGG INVESTIGATION OF VOCAL REGISTERS
This is not the place for a detailed discussion of
the cont roversi al question of vocal registers. Suf-
fice it to point out that the EGG wavef or m varies in
a characteristic manner depending on the register.
In this way, at least a qualitative document at i on can
be obt ai ned that may be useful for voi ce therapists
(14) as well as for singing t eachers.
In vocal fry or pulse register, t here can usually be
seen doubl e (so-called dichrotic) exci t at i ons and
long cl osed phases (1,10,50). In chest or modal reg-
ister, the electroglottograms normally show a some-
what rounded cl osed phase, wher eas in falsetto or
loft regi st er, t he cl osed phase t ypi cal l y is more
pointed, due to the thinning of the vocal folds and to
the very short or even insufficient cl osure of the
glottis (22,23,51-53) (Fig. 1). This distinction can
also be used to differentiate the fal set t o from the
operatic head register, even if the di fference stands
out more clearly in the phot ogl ot t ogram (51). The
ability of professional singers to egalize t he transi-
t i on be t we e n di f f er ent r egi st er s wi t hout voice
breaks or perturbations can be nicely shown in the
electroglottogram (52). Finally, by parameterizing
their EGG data, Dej onckere and Lebacq (54) have
a
C
I !
10 mS
FIG, 1. EEG waveform at different registers. Tracings of dec-
troglottograms from a normal male speaker. Increase of vocal
fold contact (glottal closure) is indicated by downward deflection
of the curves. (a) Pulse register with dichrotic wave configura~
t i ons and aperiodicity. (b) Modal regi st er, r ounded closea
phases. (c) Loft register, sharp closed phases,
Journal of Voice, Vol. 4, No. 3, 1990
ELECTROGLOTTOGRAPHY: CLINICAL APPLICATIONS 241
shown the possibility of documenting the regular
cyclic variations due to vibrato and trillo.
DOCUMENTATION OF VOICE QUALITY BY
THE ELECTROGLOTTOGRAM
Ideas about voi ce quality differ t remendousl y,
not to mention the Babyl oni an confusion prevailing
in the terminology. However , there seems to exist
widespread consent as to certain crude dimensions,
such as amount of sonori t y or occurrence of harsh-
ness, and breat hy (or leaky, hypofunctional) versus
tense (or strained, hyperfunctional) mode of phona-
tion. An obj ect i ve quantification of such dimen-
sions by analysis of the EGG waveform would cer-
tainly be helpful in clinical work, but this was re-
ported not to be possi bl e 20 years ago (55,56), and it
still does not seem possi bl e (19,25,39,57-61).
There are a number of reasons for the difficulties
in correlating the EGG waveform with dimensions
of voice quality. One of t hem is that the EGG pri-
marily informs about event s during the closed phase
of the glottal peri od. Measur es during the open
phase important for the quality of voice, like the
vibratory amplitude and the closing speed (62), can-
not at all be depi ct ed by EGG.
Based on EGG studies of more than 2,000 clinical
phoniatric cases, Sopko (22,23) tried to give a qual-
itative description of typical EGG waveforms for
different voi ce qualities. A rise of voice intensity
typically caused an i ncrease of the vi brat ory ampli-
tude in the electroglottogram. In hyperfunctional
dysphonia with strained quality of the voice, the
EGG amplitudes wer e small and often irregular,
with prolonged cl osed phases (cf. also ref. 63). Typ-
ically, for a strained or pressed quality of the voi ce
an increase of the vocal intensity did not cause the
EGG amplitudes to become larger. This was usually
the case in hypofunct i onal , l oose voi ce quality,
where the EGG wavef or m was fiat and sinosoidal at
weak intensity. However , Sopko admits that in the
evaluation of functional dysphoni as the amplitude
criterion cannot be used for comparisons bet ween
subjects because the EGG amplitude also depends
on thickness of the soft tissues in the neck. To over-
come this difficulty, Pai nt er (64) normalized his
EGG waveform so that the duration and amplitude
became about equal bef or e he classified phonations
of a professional singer into 16 different EGG wave-
form types.
In the quantitative descri pt i on of the EGG wave-
form, the relative duration of the cl osed phase has
attracted great interest as it seems to grow when the
voi ce changes from a l eaky t o a strained or pr essed
quality (Fig. 2). However , as poi nt ed out by a num-
ber of authors (10,39,46,48,65), the peak of maximal
vocal fold cont act in t he electroglottogram does not
neces s ar i l y i ndi cat e c ompl e t e gl ot t al cl os ur e.
Childers (45) even found certain breat hy voi ces to
have EGG wavef or ms similar to t hose f ound in
voi ces with compl et e glottal closure. Besi des, be-
cause of ambiguities in the definition of t he mo-
ments of closure and opening in the EGG peri od,
the duration of the cl osed phase may be difficult to
measure. In a combi ned st udy of phot ogl ot t ograms
and el ect rogl ot t ograms, the pr esent aut hor man-
aged to get accept abl e measurement s of the cl osed
phase in onl y 60% of the obs er ved el ect rogl ot -
tograms (58).
Apart from di scussi ons of whet her the moment of
glottal closure t akes pl ace at t he peak of t he EGG
waveform (66) or j ust before (1,27,47), the great est
difficulty in measuring the EGG cl osed phase is
caused by the gradual "peel i ng open" of the glottis,
so that it often becomes impossible to make out any
particular point in the cont i nuous slope of t he EGG
curve as the exact moment of glottal opening and,
t hereby, the end of the cl osed phase. A number of
authors have descri bed an irregularity or knee in the
EGG opening slope and have defined it as t he mo-
ment of opening (1,19,27,28,65,66), but the knee is
not to be found in ever y electroglottogram. Furt her-
more, irregularities of the wave slope may al so rep-
resent artefacts caused by i nst ant aneous rupt ures of
a
b
C
! !
l OmS
FIG. 2. Increasing EGG closed phase due to changes of the
voi ce quality. Normal male speaker; vocal fold cont act down-
wards. (a) leaky voice. (b) normal voi ce ( " f l ow" quality). (c)
strained or pressed voice.
Journal of Voice, Vol. 4, No. 3, 1990
242 P. KITZING
mucous bridges or by pol yps and nodules (1,10,
25,47,67).
With t he increasing ease of access to comput er
capacity, electronic analysis of waveforms has be-
come feasible even in the routine work of a voice
clinic. In the beginning, the possibility of a differ-
entiated EGG (DEGG) raised hopes for a more se-
cure identification of the opening and closing mo-
ments in the curve (10,11). Cl oser theoretical and
experimental analyses have shown, however, that
measurement s from the DEGG are not significantly
more dependabl e than t hose from the straightfor-
ward EGG (48).
Computing the quotients bet ween various seg-
ments of the wavef or m and the entire vi brat ory pe-
riod has been a common met hod of creating param-
eters to be correlated with different qualities of the
voice (68). In his monograph on EGG, Lecl use (69)
found a cl osed quotient of 0.32 in chest register and
of 0.12 for the falsetto. These dat a are in accord
with the results of a photoglottographic st udy by the
present author, where the open quotient was found
to vary bet ween 0.63 and 0.77 for low- and high-
pitched voice, respect i vel y, and bet ween 0.83 and
0.70 for weak and strong intensity (7). An increase
of the open quotient with rising pitch was found also
in EGG dat a by Rei nsch and Gobsch (70). In a study
of laryngeal paralysis, Hanson et al. (25) found the
open quotient to distinguish pathological phonation
from normal, but not to be useful for the separation
of different lesions. However , decrease in signal-
to-noise ratio oft en made the EGG signal less useful
than photoglottograms. The present author found
that EGG measurement s of the cl osed time could
differ up to about 15% from phot ogl ot t ographi c
measurement s (71). This is within the same magni-
t ude as the results of Childers et al. (45), who report
that the open (and speed) quotients can be esti-
mated from the EGG or DEGG with an average
error of - 18%, with an increased risk of error in
loud, high-pitched, and pathological voices. In the
same paper, Childers et al. (45) describe an inter-
esting criterion, t he closing vocal fol d interval,
which decreases with increasing intensity of the
voice, as was also shown by Sopko (22,23).
Trying to avoid the above-ment i oned measuring
errors, some authors have mathematically defined
some kind of global quotients that can be easily
assessed and averaged by comput ers but where the
correlation with the vi brat ory event s in the glottis
remains somewhat questionable. One such sort of
open quot i ent is the so-called bet a ([3) descri bed by
FrCkj~er-Jensen (72). The glottal cycl e is divided by
a horizontal line into a negative and a positive part
with equal areas, respect i vel y. In the comput at i ons,
the part of t he line correspondi ng t o the open phase
of the electroglottogram is divided by the entire pe-
riod, yielding the [3. In clinical applications of this
met hod, the [3 values t ended to be l ower ed in hy-
perfunctional, strained voi ces, wher eas t hey were
i ncreased in hypofunctional, l oose-qual i t y voices
(72-74).
A somewhat different quotient, the surface (S)
quotient has been descri bed by Dej onckere (75,76).
Here, the glottal cycl e is divided by a line at equal
distance from the peak and valley, and t he S quo-
tient is obtained by dividing t he areas of the result-
ing " c l os e d" part of the cycl e wi t h the " ope n"
part. Studies of pathological voi ces have shown a
statistically significant decrease of the S quotient
compared to normal voi ces (75,76).
To concl ude this section on the correlation be-
t ween the quality of the voi ce and t he EGG wave-
form, attention is drawn to the al ready mentioned
abduct i on quotient in the GLI MPES program (32-
34). Hopeful l y, this met hod of matching obtained
glottograms with simulated glottal vibrations will
make anal yses of the EGG wavef or m more useful in
the voi ce clinic. So far, in the present aut hor' s opin-
ion, the EGG wavef or m is of interest most l y for
qualitative illustrations. It should al ways be inter-
pret ed in the light of different simultaneous exami-
nations of the vocal fold vibrations (Fig. 3), such as
laryngeal st r oboscopy, as it is perfect l y possi bl e to
obt ai n normal electroglottograms from a pathologi-
cal and even cancerous larynx (Fig. 4). It is also
possi bl e t o get aberrant wavef or ms from a dys-
~
a
b
! I
10 mS
FIG. 3. Aberrant EGG waveforms due to laryngeal pathologies,
diagnosed by video-stroboscopy. Vocal fold cont act downwards.
(a) Increased closed phases due to bilateral Rei nke' s oederna,
female subject aged 65. (b) Shortened closed phases due to pare-
sis of the left recurrent laryngeal nerve, male subject aged 55.
Journal of Voice, Vol. 4, No. 3, 1990
ELECTROGLOTTOGRAPHY: CLINICAL APPLICATIONS 243
10 mS
the following sections, such measurements can be
used for different purposes, viz., to produce rec-
ords of intonation contours, to register aperiodic-
ities, and to calculate the mean and range of voice
fundamental frequency.
The analysis of intonation contours is a common
method in linguistic phonetics, and it may also be
useful in teaching the correct "accent " of foreign
languages. Also, in its clinical application, this
method is used mostly as a teaching device, show-
ing a visual display of the pitch level and intonation
pattern of spoken utterances, which can be used as
a visual feedback, e.g., in training the speech of
deaf subjects.
An intonation curve based on EGG period mea-
surements essentially consists of a string of instan-
taneous discrete values in contrast with the acoustic
fundamental frequency (F 0) of the voice, which or-
dinarily is an average estimate based on a signal
window. As pointed out by Fourcin (8,57,77), this
has the advantage that the periodicity aspect of
voice quality can also be displayed. Creaky voice
and other irregular aspects of vibration show as a
broken ragged trace in the intonation cont our,
whereas more periodic voice quality gives a smooth
contour.
Aperiodic glottal vibrations typically occur in the
beginning and end of utterances. Jentzsch, Sasama,
and Unger (60,78,79) measured the time period be-
tween the start of phonation to achievement of
steady periodic vibrations. In dysphonic patients,
they found this settling time to decrease after voice
therapy and to be increased in hypofunctional dys-
phonia. However, as shown by Kelman (80), the
time required to achieve stable periodicity also de-
pends on the vibratory frequency.
FIG. 4. EGG waveforms showing largely normal configurations
in spite of laryngeal carcinoma. Male subject aged 68. Vocal fold
contact downwards. (a) before surgery, (b) one week after taking
the specimen, and (c) photograph of the microlaryngoscopy.
functioning but healthy vocal organ (cf. also refs. 46
and 57).
INTONATION CONTOUR: SHAPE
AND SMOOTHNESS
Even if the analysis of single EGG waveforms
still seems questionable, there is general consent
that EGG is a very dependable method of measur-
ing the glottal vibratory period. As will be shown in
HARSH AND CREAKY VOICE
Aperiodic vibrations correspond to a harsh qual-
ity of voice. Instead of measuring the settling time
until the vibrations are periodic, statistical pertur-
bation measures represent a more direct way to
quantify aperiodicities. Rambaud-Pistone (81) made
a merely visual evaluation of morphological anom-
alies in the EGG waveforms, and counted, 0h av-
erage, 63% "pathological" waves in her group of 76
dysphonic patients, whereas the average was 36%
in the normal controls (n = 30). A significant vari-
ability of EGG wavef or m dat a in pathological
voices was found also by Hicks et al. (68), whereas
the measurements from normal voices showed less
Journal of Voice, Vol. 4, No. 3, 1990
244 P. KI TZI NG
dispersion (cf. also ref. 82). Measuring EGG period
pert urbat i ons (jitter) by autocorrelation linear pre-
diction was shown by Smith (83) to have a 76%
discriminating power bet ween pathological and nor-
mal voices. The discriminating power dropped to
chance level (50%) when the same analysis was per-
formed on sampled speech instead of electroglot-
tograms (83). Furt hermore, there has been shown a
cl ose relation (rs = 0.73) bet ween pert urbat i on
analysis of EGG waves and the degree of hoarse-
ness eval uat ed by auditory percept i on (37).
As already mentioned, EGG period perturbations
can be depi ct ed by the scattering of data points in
intonation cont ours based on electroglottograms.
Anot her met hod is t o pl ot s ucces s i ve per i ods
agai nst each ot her in scat t er - pl ot s, somet i mes
called "di gr ams" (8) or "bi hi st ograms" (16). This
display can be ext ended to a distributional repre-
sentation in t hree di mensi ons, where the occur-
rence rate or probability is shown by vertical excur-
sions of t he lines (2). In either of t hese displays, an
increase of pert urbat i ons can be noted in the low
frequency range when the voi ce quality is harsh or
creaky.
Analyzing frequency distributions for the occur-
rence of l ow frequency vibrations has been reported
to be an obj ect i ve measure of hoarseness, a reduc-
tion of such vibrations signaling vocal i mprovement
(59,84,85). By glottal f r equency analysis (GFA)
(18), a mi croprocessor-ai ded measurement of EGG
periods, it is possi bl e to quantify the occurrence of
l ow-frequency periods. Recent l y, Aronsson et al.
(86) showed a statistically significant correlation be-
t ween such measurement s and auditory percept ual
evaluations of harshness and creakiness by expert
listeners.
MEAN VOICE PI TCH AND RANGE
Maybe the most obvi ous way to use measure-
ments of laryngeal vi brat ory periods in the voice
clinic is to get an estimate of vocal pitch. Probabl y
it was Pawl owski and Mi t ri nowi cz-Modrzej ewska
who in 1970 wer e the first to systematically use
electroglottography to this end (87). However , even
if the voi ce F 0 depends on the vi brat ory frequency
of the vocal folds, measurement s of the vi brat ory
frequency do not al ways correspond well with ex-
pert s' estimates of the voi ce pitch, which is a psy-
choacoust i c ent i t y. The r eason is t hat l ow fre-
quency and aperiodic vibrations seem to be per-
cei ved and descri bed as changes of voi ce quality,
not so much influencing the experi ence of pitch,
which seems t o be usually based on peri odi c se-
quences of phonat i on (57). As shown in Fig. 5, au-
t omat i zed calculations of the mean vi brat ory fre-
quency typically give a l ower result than estima-
tions of mean pitch based on audi t ory percept i on.
For the measurement s to represent t he pitch cor-
rectly, vibrations representing " c r e a ky" phonat i on
first have to be eliminated. This can be accom-
plished in different ways. Chevrie-Muller (17) sug-
gest ed discarding all measur ement s differing by
more than one oct ave from the mean of the original
distribution. Fourci n (8) i nt roduced a periodicity
criterion for t he frequency histogram by increment-
ing each class of frequenci es in the histogram only
when t wo or three successi ve EGG peri ods corre-
sponded to the range of that class, so-called histo-
grams of second or third order.
In GFA (18), measurement s falling out si de the
main distribution of vi brat ory frequenci es are dis-
carded once it has dropped under a certain level (cf.
Fig. 5). The mean of the resulting " t r uncat ed" dis-
tribution has been shown to correlate very well (r =
0.98) with expert pitch estimations of more or less
pathological voices. The GFA measurement s dif-
fered from t he pitch estimations by only 2.5% on
average, the higher values pertaining to the estima-
tions. On the other hand, Comot et al. (84) have
shown the GFA values of mean frequency on the
average to be 3.75% higher t han correspondi ng
measurement s based on sonagrams.
A dependabl e met hod of measuring voi ce pitch is
especially valuable in the voi ce clinic since it has
[
[ :I :
1 ~ I !! :
I I I I I
[ I I I I ?
I I I I ! ! !
I l l l l l l !
5 I l l ! l l l ! I
I l l l l l l l !
[ r t ! l l l l f ! l l !
I * * * * ~ I I I I I I I I I I I ! I
I * * * * * * * * * * * I. I I I I I I I I I I I I I I
F 0 (SEMI TONES)
I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I I I I I I I I I STAPEL
5 I ~ 1 5 2 ~ 2 5 3 ~ 3 5 d O N n
: + * + :
FIG. 5. The st ars (*) and excl amat i on marks (!) show t he per-
cent age of occur r ences in each class of semi t ones (from 59 Hz to
750 Hz) based on 1,700 measur ement s. The mean of t hi s " r a w"
di st ri but i on is 175 Hz. Aft er di scardi ng measur ement s outside
t he mai n di st ri but i on (*), when t hi s has dr opped under a certain
level, t he resul t i ng " t r unc a t e d" di st ri but i on (!) shows a mean of
193 Hz. The pi t ch est i mat i on for t hi s voi ce by a group of trained
l i st eners was 193.8 Hz.
Journal of Voice, Vol. 4, No. 3, 1990
ELECTROGLOTTOGRAPHY: CLINICAL APPLI CATI ONS 245
been shown that correct pitch estimation is more
difficult in pathological voi ces, lacking sonority,
and at ext reme levels and wide ranges as well as at
plurimodal distributions of the speaking fundamen-
tal frequency (SFF). Besi des, it is more difficult
also when the sex of the listener differs from that of
the speaker (18).
In a recent study, Kr ook (88) found that of the
reported values of SFF in the literature >20% of t he
male voi ces and 70% of the female voi ces had been
analyzed by aid of the GFA. The average SFF val-
ues for female voi ces report ed in the literature are a
mean of 211 Hz and a range of 5.4 semitones (---1
SD) for the female voi ces and a mean of 124 Hz and
5.8 semitones for the male voices.
These values cannot be taken as a general phys-
iological norm as t hey are clearly language depen-
dent. Swedish speakers were, for instance, found to
use consi derabl y l ower speaking F o (188 Hz and 113
Hz, respectively), whereas Comot and Delaporte
using the same GFA met hod found even higher SFF
values in their French femal e speakers, viz., 222 Hz
on average (84).
The larynx reacts as an endocrinological target
organ duri ng puber t y. Peder s en et al. (89-91)
showed measurement s of the SFF to be useful as
quantifiable secondar y male sex charact eri st i cs,
highly dependent of the serum levels of sex hor-
mones. Apart from the dramatic lowering of the
male SFF during the mutation, bot h sexes, but es-
pecially females, show an age-dependent, consider-
ably smaller decrease of their SFF up to an age of
- 60- 70 years. Then t here is a certain t endency for
the SFF to rise again during senescence (84,87,92).
Besides depending on age and sex, the SFF is
also influenced by the t ype and emotional charact er
of the speech. Comparing the SFF when reading
aloud to the pitch of free speech, most authors have
reported a slightly raised SFF when reading, a find-
ing that could not be replicated by the present au-
thor. Nor did a shift of the emotional character of
the read text influence the SFF significantly. On the
other hand, changes of the entire communicative
situation or of the t ype of vocal activity were found
to have obvi ous effects on the mean pitch and range
of the voi ce (18). Provi ded such factors are kept
under control by standardized repeatable met hods
of obtaining the speech sample (like reading aloud
the same text), the mean SFF has been found to
vary within subjects by onl y 2% on average be-
tween successi ve observat i ons with a time interval
of ~2 months (18).
Such a high degree of reproducibility provi des a
basis for using measurement s of the SFF to moni t or
changes of laryngeal function in the voi ce clinic.
The most conspi cuous clinical findings are t he low-
ering of the SFF in female subj ect s by either viril-
izing hormones or smoking, and the raised pi t ch in
mal es due to puber phoni a (mut at i onal f al set t o)
(14,18,57,84). Voice t herapy for phonast heni a (vo-
cal fatigue) or functional dysphoni a was found to
cause a significant but onl y small lowering of the
mean SFF in female patients as well as a slight wid-
ening of the range (18). Examining voi ce di sorders
due to organic lesions, the GFA has pr oven espe-
cially useful in the t reat ment of Rei nke' s oedema
(polypoid degeneration), where the pathologically
l owered SFF before and the ret urn to normal levels
after surgery and voi ce t herapy could be easily doc-
ument ed (93). This usually al so had a motivating
effect on the patient to st op smoking and to engage
in voi ce therapy.
Anot her way to use measurement s of the SFF in
the voi ce clinic is in a loading test of the voi ce,
showing the capaci t y to endure vocal exertion. In
patients with functional dysphoni a, compet i ng with
70 dB white noise caused a certain rise of the mean
SFF bot h before and aft er voi ce t herapy. Aft er
reading aloud against the noise for 15 and 30 min,
there was an additional rise of the mean SFF by
@15%, but only before therapy. When the white
noise had ended, the mean SFF was still signifi-
cantly raised bef or e t herapy. In the same situation
after therapy, there was an immediate r ecover y of
the mean SFF to almost the original level (18,94).
Measurement s of the SFF range can be useful not
only in the voice clinic but also when investigating
patients with neurological and psychi at ri c distur-
bances (15,16).
APPLICATIONS OF EGG IN CERTAI N
CLINICAL DIAGNOSES
Abber t on (57) correct l y poi nt ed out that t here is
no unique relationship bet ween a pathological elec-
troglottogram and a given anatomical or physiolog-
ical condition. Nevert hel ess, several aut hors have
underlined the applicability of EGG in a numbe~ of
special diagnoses.
Bot h in a comput er model and in an in vi vo EGG
st udy of vocal fold pol yps, Childers et al. (48,67)
found t hat it was possi bl e to di fferent i at e soft,
edemat ous lesions from fi brous ones. Dej onckere et
al. (75) found their S quot i ent t o be decr eased in
Journal of Voice, Vol. 4, No. 3, 1990
246 P. KITZING
cases of vocal nodules. EGG has been successfully
used by a number of authors to get an objective
demonst rat i on of laryngeal t rauma in connect i on
with endot racheal intubation (95-98) and cricothy-
roi dot omy (99), as well as to measure t he effect of
ant i i nfl ammat ory drugs to prevent intubation le-
sions (100). Studying t he effect of laryngeal paraly-
sis on t he electroglottogram, Pedersen and Boberg
(101) did not come up wi t h concl usi ve resul t s,
whereas Hanson et al. (25) found t he glottographic
open quot i ent to distinguish paralytic phonat i on
from normal, and combined photo- and EGG to-
get her wi t h vi deost r oboscopy to be useful also
when investigating ot her t ypes of neuromuscul ar
voice i mpai rment (26). Al t ered EGG waveforms
and irregular vibrations have been found in spas-
modic dysphoni a (102) and stuttering (65,103,104).
Regarding stuttering, the conclusion of Borden et
al. (65) seems of interest; i.e., obvious temporal ir-
regularities in t he electroglottogram, such as abnor-
mal voice onset time, repetitions and prolongations,
can be explained by i mproper levels of activity
rat her than improper timing. EGG-based displays of
intonation contours have been used in t he analysis
of voice disturbances due to hearing impairment
(57). Anot her diagnostic application of EGG has
been to map t he pseudoglottis in t he upper esoph-
agus in esophageal speech, especially after trache-
oesophageal shunt operations (1,105-107).
Variations of electrical impedance in the organ-
ism obviously can be caused by ot her phenomena
than the vibrations of the larynx. For t he phoniatri-
cian and speech pathologist, an interesting applica-
tion of t he EGG is to monitor the movement s of the
soft palate as a preoperat i ve examination before
uvulopalatoplasty for severe snoring (108). Further-
more, recordi ng l ow-frequency tissue i mpedance
changes in the neck when swallowing is report ed to
have a potential for quantifying aberrant degluti-
tion, e. g. , in parkinsonian or dysphagic patients
(109).
APPLICATION OF EGG IN CLINICAL
THERAPY AND AS AN AID
As it seems, the EGG has been extensively used
for t herapy, especially in Great Britain, mostly for
dysphonias and, lately, also in connection with co-
chlear implants.
Abbert on (57) report ed that displays both of the
waveform and of EGG-based intonation contours
(VOISCOPE) were useful in a pat t ern matching
" bi of eedback" t her apy for dysphoni as and dis-
torted intonation, improving bot h percept i on and
production by establishing audi t ory and kinesthetic
awareness of the subject' s own vocal pat t erns (cf.
also refs. 14 and 105).
In children with profound hearing loss, a display
of EGG- based i nt onat i on cont our s may help to
avoid pathological speech patterns, especially high-
pitched, monot onous, and creaky voi ce (110). Pro-
foundly hearing-impaired subjects who are lipread.
ers may particularly benefit from displays of the F 0
because it is so rich in paralinguistic and prosodic
information but entirely invisible (2).
An acoustic presentation of t he intonation mel-
ody as a sinusoidal signal may be a more efficient
heari ng aid t han an ampl i fi cat i on of t he entire
speech signal (111). The same philosophy forms the
basis for using only t he F 0 signal for ext ernal elec-
t rocochl ear stimulation (57,77,109,112-114).
However , in t hese two last ment i oned applica-
tions, t here seems to be a t endency to replace EGG
by el ect roni c pitch ext ract i on f r om t he acoustic
speech signal (77,110).
CONCLUSIONS
A general conclusion on such a multifaceted topic
as t he many applications of EGG in the clinic obvi-
ously does not come easily, nor can it by any means
be scientifically grounded. But here are t he author' s
personal , ver y subjective eval uat i ons, based on
more than 15 years of almost daily use as a clinical
met hod, be it in combination with photoglottogra-
phy or stroboscopy, or as a tool to measure F0 char-
acteristics of the voice: (a) The interpretation of
measurement s of the individual EGG waveform,
and paramet ers based on t hem, seem questionable,
partly because the different phases of t he period
cannot be securel y defined and also because we do
not have sufficient knowl edge about what they
stand for physiologically and acoustically. (b) Valu-
able additional i nformat i on may be obt ai ned by
monitoring the vocal fold vibrations, when t he EGG
is used in combination with ot her met hods of inves-
tigating phonation such as vi deost roboscopy, pho-
toglottography, and inverse filtered flow glottogra-
phy. (c) Due to its paucity of overt ones, t he EGG
signal is especially well suited for measurement of
the glottal vibratory period. Such measurement s are
ext remel y useful in clinical wor k bot h for periodic-
ity analysis, as a basis for recordi ng intonation con-
tours, and to establish the characteristics of the F0
of t he voice.
Journal of Voice, Vol. 4, No. 3, 1990
E L E C T R OGL OT T OGR A P HY : C L I N I C A L A P P L I C A T I ON S 2 4 7
R E F E R E NC E S
1. Baken ILl. Electroglottography. Draft report to the Voice
Committee of the International Association o f Logopedics
and Phoniatrics. Hannover, June, 1988.
2. Fourcin A. Electrolaryngographic assessment of vocal fold
function. J Phonetics 1986;14:435-42.
3. Scherer RC, Druker DG, Titze IR. Electroglottography and
direct measurement of vocal fold contact area. In: Fujimura
O, ed. Vocal physiology: voi ce production, mechanisms
and functions, New York: Raven Press, 1988:27%91.
4. Baken RJ. Electroglottography: a broad overvi ew and be-
ginner' s introduction. Paper presented at the 18th Annual
Symposium: Care of the Professional Voice. Philadelphia,
PA, USA, June 1989.
5. Titze IR. The meaning of electroglottographic waveforms.
Paper presented at 18th Annual Symposium: Care of the
Professional Voice. Philadelphia, PA, USA, June 1989.
6. Rothenberg M. A new electroglottograph. Paper presented
at the 18th Annual Symposium: Care of the Professional
Voice. Philadelphia, PA, USA, June 1989.
7. Kitzing P, Sonesson B. A photoglottographical study of the
female vocal folds during phonation. Folia Phoniatr (Basel)
1974;26:138-49.
8. Fonrcin AJ. Laryngographic assessment ofphonat ory func-
tion. In: Ludl ow C, O' Connel l H, eds. Proceedings of the
conference on the assessment of vocal pathology. ASHA
Reports 1981 ; 11:116-27.
9. FrCkj~er-Jensen B. Personal communication, 1989.
10. Childers D, Larar JN. Electroglottography for laryngeal
funct i on assessment and speech analysis. I EEE Trans
Bi omed Eng 1984;31:807-17.
11. Hanson D, Gerratt BR, Watd PH. Glottographic measure-
ment of vocal dysfunction. A preliminary report. Ann Otol-
rhinollaryngol 1983;92:413-20.
12. Krisnamurthy AK, Childers DG. Two-channel speech anal-
ysis. I EEE Transactions on Acoustics, Speech, and Signal
Processing 1986;34:730-43.
13. Trapp TK, Berke GS. Photoelectric measurement of laryn-
geal paralyses correlated with videostroboscopy. Laryngo-
scope 1988;98:486-92.
14. Carlson E. Electrolaryngography as a tool for routine voi ce
assessment and t herapy. The Voice Research Soci et y
Newsletter, 1988;2:14-29.
15. Guidet C, Chevrie-Muller C. Computer analysis of prosodic
and electroglottographic parameters in diagnosis of patho-
logic voice. In: Winkler P, ed. Investigations of the speech
process, quantitative linguistics. Bochum; Studienverlag
Dr. N. Brockmeyer, 1979:19.
16. Guidet C, Chevrie-Muller C. M6thode de traitement du sig-
nal 61ectroglottographique. Application au diagnostic au-
tomatis6 des troubles de la phonation. Innov Tech Biol Med
1983;4:617-35.
17. Chevrie-Muller C, Dordain M, Decante P. Analyse auto-
matisEe des param~tres acoustiques de la parole. In: Union
of the European Phoniatricians, I V Congress. Papers and
abstracts. Wroclaw. 1975:90-3.
18. Kitzing P. Glottografisk rekvensindikering [thesis]. Malm6,
1979. Glottal frequency analysis [(condensed version in En-
glish). Paper presented at The Society for Voice Research,
Royal Society of Medicine, London, September 1986.]
19. Anastaplo S, Karnell MP. Synchronized videostroboscopic
and electroglottographic examination of glottal opening. J
Acoust Soc Am 1988;83:1883-90.
20. Karnell MP. Synchronized videostroboscopy and electro-
glottography. J Voice 1989;3:68--75.
21. Roch JB. Personal communication, 1989.
22. Sopko J. Die klinische Phoni at ri e--Ergebni sse der ver-
gleichenden Untersuchungen des gesunden und kranken
Kehlkopfes beim Menschen mittels Photographie, Laryn-
gostroboskopie und El ekt rogl ot t ographi e. Habilitations-
schrift (Thesis). Basel: Uni versi t y of Basel, 1983;1-126.
23. Sopko J. Zur Objektivierung der Stimmlippenschwingun-
gen mittels synchroner elektroglottographischer und stro-
bos kopi s cher Unt er s uchung. Spr ache- St i mme- GehOr
1986;10:83-7.
24. Gerratt BR, Hanson DG, Berke GS. Laryngeal configura-
tion associ at ed with gl ot t ogr aphy. Am J Ot ol aryngol
1988;9:173-9.
25. Hanson DG, Gerratt BR, Karin RR, Berke GS. Giotto-
graphic measures of vocal fold vibration: an examination of
laryngeal paralysis. Laryngoscope 1988;98:541-9.
26. Hanson DG, Ward PH, Gerratt BR, Berci G, Berke GS.
Diagnosis of neuromuscular voi ce impairment. In: Decker
BC. Geriatric otorhinolaryngol. 1989:71-8.
27. Kitzing P. Methode zur kombinierten photo- und elektro-
glottographischen Registrierung von Stimmlippenschwing-
ungen. Folia Phoniatr (Basel) 1977;29:249-60.
28. Kitzing P, Lf f qvi s t A. Clinical application of combi ned
electro- and photoglottography. Paper presented at t he 17th
Congress of the International Association of Logopedi cs
and Phoniatrics. Copenhagen, 1977.
29. Rothenberg M. Some relations bet ween glottal air fl ow and
vocal fold contact area. Proceedings of the Conference on
the Assessment o f Vocal Pathology. Bethesda, MD: NIH,
1979;1-19.
30. Berry RJ, Epstein R, Fourci n A J, Freeman M, MacCurtain
F, Noscoe N. An obj ect i ve analysis of voi ce disorders: Part
One and Two. Br J Disord Commun 1982;17:67-83.
31. Lecluse FLE. Personal communication, 1989.
32. Scherer RC, Titze IR. The abduction quotient related to
vocal quality. J Voice 1987;1:246-51.
33. Scherer RC, Gould WJ, Titze IR, Meyers AD, Sat al offRT.
Preliminary eval uat i on of sel ect ed acoustic and giotto-
graphic measures for clinical phonatory function analysis, J
Voice 1988;2:230--44.
34. Titze 1R. Parameterization of the glottal area, glottal flow,
and vocal fold contact area. J Acoust Soc Am 1984;75:570-
80.
35. Ferrero FE. L' elettroglottografia. In: Croatto L, ed. Trat-
tato di foniatria e logopedia, Vol 4. Padova: Editrice " La
Garangol a" 1987;109-62.
36. Askenfelt A, Gauffin J, Sundberg J, Kitzing P. A compar-
ison of contact microphone and electroglottograph for the
measurement of vocal fundamental frequency. J Speech
Hear Res 1980;23:258-73.
37. Haji T, Horiguchi S, Baer T, Gould WJ, Wilbur J. Fre-
quency and amplitude perturbation analysis of electroglot-
tograph during sustained phonation. J Acoust Soc Am
1986;80:58-62.
38. Hol m C. L' 6vol ut i on de la phonation de la premi ere en-
fance ~t la pubert6: une 6tude 61ectroglottographique. J Fr
Otorhinolaryngol 1971;20:437-40.
39. Baken RJ. Clinical measurement o f speech and voice. Lon-
don: Taylor & Francis, 1987.
40. Cavallo SA, Baken RJ. Prephonatory laryngeal and chest
wall dynamics. J Speech Hear Res 1985;28:79-87.
41. Smith S. Research on the principle of electroglottography.
Folia Phoniatr (Basel) 1981 ;33:105-14.
42. FrCkj~er-Jensen B. Interpretation of electro-glottogral~hic
recordings. Ann Rep Inst Phon Univ Copenhagen 1982;16:
241-5.
43. Gilbert HR, Potter CR, Hoodi n R. Laryngograph as a mea-
sure of vocal fol d cont act area. J Speech He ar Res
1984;27:173-8.
44. Berke GS, Moore DM, Hant ke DR, Hanson DG, Gerratt
BR, Bnrstein F. Laryngeal modeling: theoretical, in vitro,
in vivo. Laryngoscope 1987;98:871-81.
Journal of Voice, Vol. 4, No. 3, 1990
248 P. K I T Z I N G
45. Childers DG, Hicks DM, Moore GP, Eskenazi L, Lalwani
A. Electroglottography and vocal fold physiology. J
Speech Hear Res (in press).
46. Childers DG, Smith AM, Moore GP. Relationships be-
tween electroglottograph, speech, and vocal cord contact.
Folia Phoniatr (Basel) 1984;36:105-18.
47. Childers DG, Krishnamurthy AK. A critical review ofelec-
troglottography. Crit Rev Biomed Eng 1985;12:131-61.
48. Childers DG, Hicks DM, Moore GP, Alsaka YA. A model
for vocal fold vibratory motion, contact area, and the elec-
troglottogram. J Acoust Soc Am 1986;80:1309-20.
49. Childers DG. Laryngeal function: objective-perceptual
classification. Final report (grant number: CMS 5 R01
NS17078-06). Washington, DC: National Institutes of
Health, June 1988.
50. Kitzing P, L0fqvist A. Evaluation of voice therapy by
means of photoglottography. Fol i a Phoni at r (Basel)
1979;31:103-9.
51. Kitzing P. Photo- and electroglottographical recording of
the laryngeal vibratory pattern during different registers.
Folia Phoniatr (Basel) 1982;34:234--41.
52. Roubeau B, Chevrie-Muller C, Arabia-Guidet C. Electro-
glottographic study of the changes of voice registers. Folia
Phoniatr (Basel) 1987;39:280-9.
53. Schutte HK, Seidner WW. Registerabhgngige Diffe-
renzi erung von El ekt rogl ot t ogrammen. Spr ac he -
Stimme-GehOr 1988;12:59--62.
54. Dejonckere PH, Lebacq J. Differences in physiological
mechanisms for producing vocal vibrato, trillo and tremolo.
Act a Phoniatrica Latina (in press).
55. Loebell E. Ober den klinischen Wert der Elektroglotto-
graphie. Arch Klin Exp Ohr-, Nas- Kehl k- Hei l k 1968;
191:760M.
56. Loebell E. Neue Ergebnisse der Elektroglottographie
(EGG). In: Bustamante Gurria A, ed. Ot o- rhi no-
laryngology, Proceedings o f the ninth international con-
gress, Mexi co, 1969. Amsterdam: Excerpta Medica,
1970:272-8.
57. Abberton E, Fourcin A. Electrolaryngography. In: Code C,
Ball M, eds. Experimental clinical phonetics. London:
Croom Helm, 1984:62-78.
58. Kitzing P. Glottography, the electrophysiological investiga-
tion of phonatory biomechanics. Act a Otorhinolaryngol
Belg 1986;40:863-78.
59. Neil WF, Wechsler E, Robinson JMP. Electrolaryngogra-
phy in laryngeal disorders. Clin Otolaryngol 1977;2:33-40.
60. Unger E. Die Erfassung und Bewertung objektiver Para-
meter des Stimmapparates mittels der Elektroglottographie
und der Pneumotachographie [thesis]. Erfurt: Medizi-
nische Akademie, 1979.
61. Unger E, Unger H, Tietze G. Stimmuntersuchungen mittels
der elektroglottographischen Einzelkurven. Folia Phoniatr
(Basel) 1981;33:168-80.
62. Fant G. Glottal source and excitation analysis. Speech
Transmission Laboratory Quarterly Progress and Status
Report 1979;1:85-107.
63. van Michel C. Morphologie de la courbe glottographique
darts certains troubles fonctionels du larynx. Folia Phoniatr
(Basel) 1967;19:192-202.
64. Painter C. Electroglottogram waveform types. Arch Otorhi-
nolaryngol 1988;245:116-21.
65. Borden GJ, Baer T, Kay Kenney M. Onset of voicing in
stuttered and fluent utterances. J Speech Hear Res
1985;28:363-72.
66. Baer T, Lffqvist A, McGarr NS. Laryngeal vibrations: A
comparison between high-speed filming and glottographic
techniques. J Acoust Soc Am 1983;74:1304-8.
67. Childers DG, Alsaka YA, Hicks DM, Moore GP. Vocal
fold vibrations in dysphonia: model vs. measurement, j
Phonetics 1986;14:42%34.
68. Hicks DM, Bae KS, Childers DG, Moore GP. Objective
confirmation of perceptual evaluation of vocal disorders.
Paper presented at the AS HA Convention. New Orleans,
LA, USA, 1987.
69. Lecluse FLE. Elektroglottografie. An experimental study
of the electrical impedance for the male human larynx (The-
sis). Utrecht: University of Rotterdam, 1977;1-181.
70. Reinsch M, Gobsch H. Zur quantitativen Auswertung elek-
troglottographischer Kurven bei Normalpersonen. Folia
Phoniatr (Basel) 1972;24:1-6.
71. Kitzing P. Simultaneous photo- and electroglottographic
measurements of voice strain. In: Titze IR, Scherer RC,
eds. Vocal fold physiology. The Denver Center for the Per-
forming Arts, 1983:221-9.
72. FrCkj~er-Jensen B. Can glottography be used in the clinical
practice? Paper presented at the XIX Congress of the In-
ternational Association of Logopaedics and Phoniatrics.
Edinburgh, 1983.
73. Hacki T. Beitrag zur Klassifizierung von Glottis-
Dysfunktionen mit Hilfe der Elektroglottographie. Folia
Phoniatr (Basel) 1989;41(1):43--8.
74. Thyme K, FrCkj~er-Jensen B. Acoustic changes caused by
speech training compared with changes in electro-
glottographic registrations. Paper presented at the Con-
gress of Union of European Phoniatricians. Praha, 1982.
75. Dejonckere PH, Lebacq J. Electroglottography and vocal
nodules--an attempt to quantify the shape of the signal.
Folia Phoniatr (Basel) 1985;37:195-200.
76. Dejonckere PH. Quantitative parameters in EGG. Paper
presented at the XVth Congress of the Union of European
Phoniatricians. Erlangen, 1988.
77. Abberton E, Fourcin AJ, Rosen S, et al. Speech perceptual
and productive rehabilitation in electro-cochlear stimula-
tion. In: Schindler RA, Merzenich MM, eds. Cochlear im-
plants. New York: Raven Press, 1985:527-37.
78. Jentzsch H, Sasama R, Unger E. Elektroglottographische
Untersuchungen zur Problematik des Stimmeinsatzes bei
zusammenh~ingendem Sprechen. Folia Phoniatr (Basel)
1978 ;30:59-66.
79. Jentzsch H, Unger E, Sasama R. Elektroglottographische
Verlaufs-kontrollen bei Patienten mit funktionellen
Stimmst6rungen. Folia Phoniatr (Basel) 1981 ;33:234-41.
80. Kelman AW. Vibratory pattern of the vocal folds. Folia
Phoniatr (Basel) 1981 ;33:73-99.
81. Rambaud-Pistone E. Place de l'6tude instrumentale dans le
bilan vocal. Bulletin d'audiophonologie 1984;17:41-58.
82. Dejonckere PH. Les troubles du rythme glottique. Acta
Otorhinolaryngol Belg 1985;39:837--43.
83. Smith AM, Childers DG. Laryngeal evaluation using fea-
tures from speech and the electroglottograph. I EEE Trans
Bi omed Eng 1983;30:755-9.
84. Comot C, Delaporte C. l~tude du fondamental laryng6 dans
la voix parl6e chez l'adulte [th~se]. Facult6 "Techniques de
R6adaptation," Lyon, 1987.
85. Wechsler E. A laryngographic study of the voice disorders.
Br J Disord Commun 1977;12:9-22.
86. Aronson A, Henriksson I, Sandstedt E. J~imf6relse mellan
elektroglottografisk analys av r6stkvalitet och perceptuell
r6stbed6mning f6re och efter r6stterapi [thesis]. Institu-
tionen f6r logopedi och foniatri, G6teborgs universitet,
1986.
87. Pawlowski Z, Mitrinowicz-Modrzejewska A. The results of
vocal organ investigation of the deaf. FoUa Phoniatr
(Basel) 1970;22:434--44.
88. Krook MIP. Speaking fundamental frequency characteris-
tics of normal swedish subjects obtained by glottal fre-
quency analysis. Folia Phoniatr (Basel) 1988;40:82-90.
Journal of Voice, Vol. 4, No. 3, 1990
ELECTROGLOTTOGRAPHY: CLI NI CAL APPLI CATI ONS 249
89. Pedersen MF, Kitzing P, Krabbe S, Heramb S. The change
of voice during puberty in 1l to 16 years old choir singers
measured with electroglottographic fundamental frequency
analysis and compared to other phenomena of puberty.
Acta Otolaryngol [Suppl] (Stockh) 1982;386:189-92.
90. Pedersen MF, Mller S, Krabbe S, Munk E, Bennet P,
Kitzing P. Change of voice in puberty in choir girls. Acta
Otolaryngol [Suppl] (Stockh) 1984;412:46-9.
91. Pedersen MF, Mller S, Krabbe S, Bennet P. Fundamental
voice frequency measured by electroglottography during
continuous speech. A new exact secondary sex character-
istic in boys in puberty. Int J Ped Otorhinolaryngol
1986;11:21-7.
92. Chevrie-Muller C, Perbos J, Guidet C. Automated analysis
of the electroglottographic signal: application to the study
of phonation of the elderly. BullAudiophonol 1983;16:121-
44.
93. Kitzing P. Phoniatric aspects of microlaryngoscopy. Acta
Phoniatrica Latina 1987;9:31-41.
94. Kitzing P, _Akerlund L. Anderungen der Sprechstimmlage
im Stimmbelastungstest. Paper presented at the II. Kom-
munikationsmedizinische Tage. Bad Rappenau, Germany,
April 1988 (in press).
95. Gleeson MJ, Fourcin AJ. Clinical analysis of laryngeal
trauma secondary to intubation. J R Soc Med 1983;76:928-
32.
96. Lesser THJ, Williams RG, Hoddinott C. Laryngographic
changes following endotracheal intubation in adults. Br J
Disord Commun 1986;21:239-44.
97. Lesser THJ, Lesser PJA. Laryngeal trauma vs. length of
intubation. J Laryngol Otol 1987;101:1165-7.
98. Lesser THJ, William G. Laryngographic investigation of
postoperative hoarseness. Clin Otolaryngol 1988;13:37-42.
99. Gleeson MJ, Pearson RC, Armistead S, Yates AK. Voice
changes following cricothyroidotomy. J Laryngol Otol
1984;98:1015-9.
100. Mazzarella B, Macarone Palmieri A, Mastronardi P, et al.
Benzydamine for the prevention of pharyngo-laryngeal pa-
thology following tracheal intubation. Int J Tissue React
1987;9:121-9.
101. Pedersen MF, Boberg A. Examination of voice function of
patients with paralysis of the recurrent nerve. Acta Oio-
laryngol 1973;75:372--4.
102. Chevrie-Muller C. Arabia-Guidet C, Pfauwadel MC. Can
one recover from spasmodic dysphonia? Br J Disord Com-
mun 1987;22:117-28.
103. Chevrie-Muller C. t~tude du fonctionnement laryng6 chez
les b~gues par la m6thode gtotto-graphique. Rev Laryngol
Otol Rhinol (Bord) 1964;9-10:763-74.
104. Conture EG, Rothenberg M, Molitor R. Electroglotto-
graphic observations of young stutterers' fluency. J Speech
Hear Res 1986;29:384--96.
105. Lecluse FLE, Tiwari RM, Snow GB. Electroglottographic
studies of staffieri neoglottis. Laryngoscope 1981 ;91:971-5.
106. Singh W. Clinical application of electrolaryngograph for
speech rehabilitation in near-total laryngectomy with myo-
mucosal valved neoglottis. J Laryngol Oto11988;102:335-6.
107. Tiwari RM, Snow GB, Lecluse FLE, Greven AJ, Bloot-
hooft G. Observations on surgical rehabilitation of the
voice after laryngectomy with Staffieri's method. J Laryn-
gol Otol 1982;96:241-50.
108. Chouard CH, Meyer B, Chabolle F. The Velo-imped-
ancemetry. Acta Otolaryngol (Stockh) 1987;103:537-45.
109. Sorin R, McClean MD, Ezerzer F, Meissner-Fischbein B.
Electroglottographic evaluation of the swallow. Arch Phys
Med Rehabil 1987;68:232-5.
110. Abberton E, Fourcin AJ, Hazen V. Phonological compe-
tence with profound hearing loss. In: Taylor IG, ed. The
education of the deaf. Current perspectives. Volume I.
Croom Helm, 1985:44-55.
111. Rosen S, Walliker JR, Fourcin A, Ball V. A microproces-
sor-based acoustic hearing aid for the profoundly impaired
listener. J Rehabil Res Develop 1987;24:239--60.
112. Abel SM, Tse S, Kunov H, Nedzelski J, Harrison R. The
use of a temporary extracochlear electrode in preoperative
testing of permanent implant condidates. Scand Audiol
1986;15:197-205.
113. Fourcin A, Douek E, Moore B, Abberton E, Rosen S, Wal-
liker J. Speech pattern element stimulation in electrical
hearing. Arch Otolaryngol 1984;110:145-53.
114. Fourcin AJ. Phonetics and cochlear prostheses. Audiology
in Practice 1987;4:1-4.
Journal of Voice, Vol. 4, No. 3, 1990